Research map / Detection and segmentation / CNN features
Learning Where To Learn in Cross-View Self-Supervised Learning
Lang Huang, Shan You, Mingkai Zheng, Fei Wang, Chen Qian, Toshihiko Yamasaki
Check your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add LEWEL alignment learning to BYOL | BYOL · ImageNet-1K linear classification, 100 epochs | Acc@1: 1.3% points | Improved |
| Add LEWEL alignment learning to BYOL | BYOL · ImageNet-1K semi-supervised classification, 1% labels, 200 pre-training epochs | Acc@1: up to 1.3% points | Improved |
| Add LEWEL alignment learning to MoCov2 | MoCov2 · Pascal VOC 07+12 detection, 100 epochs | AP: 0.4% points | Improved |
| Add LEWEL alignment learning to BYOL | BYOL · Pascal VOC 12 semantic segmentation, 100 epochs | mIoU: 0.7% points | Improved |
| Add aligned loss to global loss (ablation) | LEWEL_M · ImageNet-100 linear classification | Acc.: 79.5 -> 81.0 | Improved |
| Add aligned loss to global loss (ablation) | LEWEL_M · Pascal VOC 12 segmentation | mIoU: 61.6 -> 62.7 | Improved |
| Couple global and aligned projections through weight sharing (ablation) | LEWEL_M with global and aligned losses · ImageNet-100 linear classification | Acc.: 81.0 -> 82.1 | Improved |
| Couple global and aligned projections through weight sharing (ablation) | LEWEL_M with global and aligned losses · Pascal VOC 12 segmentation | mIoU: 62.7 -> 63.4 | Improved |
| Replace learned alignment maps with random maps (ablation) | LEWEL_M · ImageNet-100 linear classification | Acc.: 82.1 -> 79.8 | Hurt |
| Replace learned alignment maps with random maps (ablation) | LEWEL_M · Pascal VOC 12 segmentation | mIoU: 63.4 -> 61.9 | Hurt |
| Increase MoCov2 projection dimension from 128 to 512 (ablation) | MoCov2 · ImageNet-100 linear classification | Acc.: 79.5 -> 80.2 | Improved |
| Increase MoCov2 projection dimension from 128 to 512 (ablation) | MoCov2 · Pascal VOC 12 segmentation | mIoU: 61.6 -> 61.5 | No clear effect |
Closest papers by meaning
- Semantic Concentration for Self-Supervised Dense Representations Learning (TPAMI 2025 · similarity 0.87)
- Harnessing small projectors and multiple views for efficient vision pretraining (NeurIPS 2024 · similarity 0.87)
- Joint‑Embedding vs Reconstruction: Provable Benefits of Latent Space Prediction for Self‑Supervised Learning (NeurIPS 2025 · similarity 0.86)
- Self-Supervised Learning from Structural Invariance (ICLR 2026 · similarity 0.86)
- CG-SSL: Concept-Guided Self-Supervised Learning (NeurIPS 2025 · similarity 0.86)
- Probing the Mid-level Vision Capabilities of Self-Supervised Learning (CVPR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
